OpenAI has launched ChatGPT for Teens, a version of the chatbot designed to support learning while applying stronger built-in safety protections for under-18 users.
Users are automatically placed into the teen experience if OpenAI's age-prediction system estimates they are under 18, or if they state their age as between 13 and 17. The product builds on earlier work including parental controls, an under-18 model specification and a "Teen Safety Blueprint."
The learning-focused features include Study Mode, which uses guiding questions rather than direct answers to walk teens through material, alongside quizzes, learning visualisations and a new "responsible homework reminder" designed to redirect teens away from shortcutting assignments. Parents and teens can also set defined "Study Hours" during which Study Mode runs by default.
Safety protections are on by default and cover higher-risk areas including self-harm, violence, eating disorders, dangerous activities and explicit content. Parents with linked accounts can set quiet hours, manage settings and receive notifications in limited high-risk situations, including a new alert category for eating disorders. OpenAI's updated under-18 model guidance also bars romantic language, discourages emotional dependence, and stops the model implying it has feelings or consciousness.
OpenAI has also announced a partnership with CodeAI aimed at helping teens and teachers understand how AI systems work, and says it will begin publishing under-18 evaluation results from its model system cards covering categories such as self-harm and sexual content.
